УДК621,793
I. Gedzavicius, A.V. Valiulis
RESEARCH OF THE THERMAL SPRAY PROCESS OF WIRE ARC SPRAYING
Department o f Materials Science and Welding, Vilnius Gediminas Technical University, Vilnius, Lithuania
Introduction
Thermal spray has been used for about 40 years throughout all the major engineering industry sectors for component protection and reclamation. Arc spraying is the highest productivity thermal spray process. Recent equipment and process developments have improved the quality and expanded the potential application range for thermally sprayed coatings. Thermal spray is a well established technology for applying wear and corrosion resistant coatings in many key industrial sectors, including aerospace, automotive, power generation, petrochemical and offshore. In recent years, improvements to equipment and material quality have enhanced the technical credibility o f the thermal spray processes, leading to a significant growth in new markets, e.g., biomedical, dielectric and electronic coatings.
A DC electric arc is struck between two continuous consumable wire electrodes which form the spray material. Compressed gas (usually air) atomises the molten material drops into fine droplets and propels them towards the substrate. The process is simple and can be operated either manually or in an automated manner. It is possible to spray a wide range o f metals, alloys and metal matrix composites (MMCs). In addition, a limited range o f cermet coatings (with tungsten carbide) can also be sprayed in cored wire form, where the hard ceramic phase is packed into a metal core as a fine powder.
I o f arc spray gun
569
1Ъе aim of this study is to tty to optimise the atomizing gas velocity perfomiing small changes on the TAFA9000 gunnozde geometry, using fixed thennal sprayparameteis. An increase in the particle velocity, providing improvements in the coating properties is expected [1].
NUMERICAL STUDY OF THE ATOMISING GAS JET
The PHOENICS commercial CFD (Computational Fluid Dynamics) code was used in a preliminary step in order to test the effect o f small changes made on the original design of the TAFA 9000 gun nozzle geometry [2]. The original gun exit is composed o f a converging nozzle exhibiting a 6 mm exit diameter. The meeting point o f the wires (where the electric arc is formed) is situated just at the center o f the exit area. The changes that were made on nozzle geometry and were tested included additional extension of the original nozzle.(Fig.
2). Di^erent lengths and different diverging angles were tested. Simple two-dimensional computations were performed and the effect o f the electric arc (150A; 30V) was taken into account by adding a volumetric heat source to the energy equation. This simplified approach was expected to lead to the first tendency (qualitative comparison) o f the effect o f these small changes. FinaHy, six different new nozzle exit designs were tested. Figure 2 presents a view of these different nozzle exit designs. In each case, the results were compared to those obtained for the original geometry (Figure 2-a). More precisely, criteria such as- the velocity magnitude and the jet divergence in the near exit region were retained. Figure 2-b incoiporates a progressive change in the converging angle. Figures 2-c to 2-e show different lengths o f constant area nozzle extensions whereas geometries on Figures 2-f and 2-g were built up with a slightly diverging extension.
Fig. 2. View o f the tested diverging nozzles
Finally, one of the modified designs was selected on the basis o f the nnmencal modeling results and the corresponding nozzle was machined in order to be tested on the experimental way. The selected design has a 3 mm constant area extension (Figure 1 -d). ТЪе other designs were modeled to provide either a lower gas velocity or a larger jet divergence (if compared to the chosen one).
CHARACTERIZATION OF THE MATER JETS
Following this first preliminary modeling step, the corresponding nozzle (constant area 3 mm extended nozzle) was machined and tested Some experiments were performed tor two different type wires (TAFA 38T steel wire and ТАРА 95MXC stainless steel cored wire). Figure 3 and 4 present photographs of the jet for the TAFA 95MXC wire, for the original and modified nozzles respectively. The other parameters were kept unchanged (electric arc current -150A, voltage- 30 V and a gas flow rate o f 130 m^/h). The photographs were performed using a numerical camera and visual references allowed a calibration of the photos. In spite o f the problem concerning the matter jet motion, the jet divergence was found to be lower for the modified nozzle geometry (12®±1 against 16®±1). Additionally, the matter distribution at the spraying distance (150 mm) is more expanded in comparison to the original geometry.
Fig.S. Jet o f original nozzieFig. 4. J e t o f m odified nozzle
PARTICLE SIZE DISTRIBUTION
After these experiments, it was decided to perform investigation o f the particle jet.
For the new experiments, the thermal spray parameters were the same for the two designs (previous ones) so that only the effect o f the nozzle geometry was taken into account. In a preliminary step, particles firom the spray o f two different nozzle designs were collected Into water container and a sieve analysis was performed using a LASER granulometer in order to allow a better calibration o f the particle size distribution. Additionally, SEM micrographs of the collected particles were also made and analysed.
OiauKtcr of the poiticics ( дш)
Fig. 5. Diam eter o f the particles f o r the original and m odified nozzles
Figure 5 presents the particle size distribution taken from the sieve analysis for the two wires and the two different nozzles. The size o f particles is equally distributed on thd logarithmic scale. For the steel 38T wire, the mean particle size was 26.1 ?m for the original nozzle and 22.2 ?m using the modified one. In case o f the stainless steel 95MX€ wire, the mean particle size was smaller. The mean particle size 18.7 ?m for the original nozzle and about 14 ?m for the nozzle o f the modified geometry.
Consequently, the mean particle size of the collected particles, was about 15% lower for the 3 8T wire and 25% lower for the 95MXC wire using the modified nozzle. The sprayed particles o f wire 35T are smaller than of cored wire 95MXC. This difference may be explained by different chemical compositions and surface tension o f the liquid droplets in flight.
CONCLUSIONS 4
1. Computational Fluid Dynamics models can predict the influence o f nozzle geometry on flows o f jet and heat transfer. Also it helps to choose optimal nozzle configuration.
2. Jet flow from the modified geometry gun nozzle is converged and forms droplets o f smaller size. This allowed to predict that porosity o f sprayed coatings and the consuming o f spray material will decrease, while the adhesion and Young’s modulus will increase.
3. The investigation o f arc spray with modified nozzle must be continued with sprayed coatings made fiom different materials and the detailed examination o f mechanical and physical properties o f coatings must be carried out with the aim to set the optimal relation of gun nozzle geometry and sprayed coating properties.
REFERENCES
1. Edited by Christian Coddet «Thermal Spray Meeting the challenges o f the 2 И Century», proceedings o f the 15^ International Thermal Spray Conference 15-29 May 1998 Nice, France. Volumes 1,2 p.l730.
2. R. Bolot, R. Bonnet, G. Jandin, C. Coddet “Application o f CAD to CFD for the Wire Arc Spray Process”, Thermal Spray 2001: New Surfaces for a New Millennium, (Ed») ( Bemdt, К A . Khor, and E. E Lugscheider, Published by ASM International, Materials Park, Ohio, USA, 2001, pp. 889-894.
3 . N.A. Hussary, and J. Heberlein, “Investigations o f Arc Behavior and Particle I'ormation in the Wire Arc Spray Process Using ffigh Speed Photography”, Thermal Spray:
Surface Engineering via Applied Research, Ed. C.C. Bemdt, Pub ASM Int., Materials P a ± , o n , 2000, pp. 737-742.
4. L.Leblanc, P. Gougeon, C. Moreau “Investigation o f die Long-Term Stability o f Plasma Spraying by Monitoring Characteristics o f the Sprayed Particles”, Thermal Spray:
A united Forum for Scientific and Technological Advances, C.C. Bemdt (Ed), Published by ASM International, Materials Park, Ohio, USA, 1997, pp. 567-575.
5. Takayuki Watanabe, Tadayuki Sato and Atsushi Nezu «Electrode phenomena Investigation o f wire arc spraying for preparation o f Ti-Al intermetallic compounds». Thin Solid Films, March 2002, p.p 98-103
6. Moreau C., Cielo P. et Lamontagne M., «Flattening and solidification o f thermally sprayed particles», Journal o f Thermal Spray Technology, Vol 1, No 4.1992. pp. 317 -323.